Research on Association Rules Mining Based-On Ontology in E-Commerce

Commercial activities carried out through the Internet become more and more popular. And lots of transaction logs are generated, which means we can gain useful information by data mining. Thereby, association rules mining is very valuable in e-commerce. But there are some problems of existing association rules mining systems. The existing traditional approaches can´t solve these problems very well. In order to solve these problems better, this paper proposes association rules mining based-on ontology, and mainly researches the following three parts during data mining: (1) methods of ontology construction and principles of commodity classification; (2) simplifying R-interesting according to actual situations; (3) implementing association rules mining based-on ontology by improved Apriori. Moreover, this paper tests the improved algorithm using FoodMart2000, Java as the development language and Jena as the ontology engine, finishes the whole process of mining, and verifies the validity of the algorithm by the example of the database.